Figure 3.
Dynamic changes in Jak2V617F/+Trp53−/− HSPCs occur during leukemogenesis. (A) Heat map of top representative genes (top 30 genes) for each cluster from single-cell expression profiles. Gene expression profiles of selected marker genes were used to assign cell classification. Columns represent individual cells; rows represent genes. (B) Average expression levels of conventional surface marker genes Ptprc (CD45), Cd34 (CD34), Fcgr3 (CD16), Fcgr2b (CD32), Tfrc (CD71), and Itga2b (CD41) for single cells in each cluster. Cell clusters were labeled and colored by cell type names: 1, hematopoietic stem cells-1 (HSC-1); 2, HSC-2; 3, Pre-GMP; 4, GMP-1; 5, GMP-2; 6, GMP-3; 7, megakaryocyte-erythroid progenitor (MkE-P); 8, EryP-1; 9, EryP-2; 10, erythroid blast-1 (EryB-1); 11, EryB-2; 12, leukemia-1; 13, leukemia-2; 14, megakaryocyte progenitor-1 (MkP-1); 15, MkP-2; 16, megakaryocyte blast (MkB); and 17, B-cell progenitor (B-cell-P). Dot plot displays average gene expression of surface marker genes for identification of a cluster cell type. The color key from white to red indicates low to high average gene expression level, respectively. The dot size indicates percentage of cells expressing a certain marker. (C) Unsupervised graph-based clustering of scRNA-seq data set projected onto a t-distributed stochastic neighbor embedding (t-SNE) plot displaying distinct clusters. A total of 29 029 cells were clustered in the left panel. Each point represents an individual cell. Cell clusters were labeled and colored according to cell type (HSPC subpopulations and leukemic population). In the right panel, t-SNE plot displays subpopulations of HSPCs and leukemic population in a WT, JVF, a pre-PEL JVFP−/−, and a PEL JVFP−/− mouse. (D) Heat map of hierarchical clustering of Pearson correlation coefficients between clusters. All coefficients based on the average normalized scRNA-seq data in each cluster. (E) Enrichment analysis of Gene Ontology (GO) and pathways associated with upregulated genes in leukemia cells relative to erythroid progenitors and blasts. (F) Inferred CNA profiles from scRNA-seq data. Chromosomal gains (red) and deletion (blue) are inferred by averaging expression over 100-gene stretches on the respective chromosomes (columns) across the single cells (rows). Top: cells from a WT mouse as reference cells not expected to contain CNVs. Bottom: cells tested for CNAs relative to the reference cells. Zoom in of CNA profiles for leukemia, EryP, and EryB groups on chromosome 4, 6, 10, and 15 (right).

Dynamic changes in Jak2V617F/+Trp53−/− HSPCs occur during leukemogenesis. (A) Heat map of top representative genes (top 30 genes) for each cluster from single-cell expression profiles. Gene expression profiles of selected marker genes were used to assign cell classification. Columns represent individual cells; rows represent genes. (B) Average expression levels of conventional surface marker genes Ptprc (CD45), Cd34 (CD34), Fcgr3 (CD16), Fcgr2b (CD32), Tfrc (CD71), and Itga2b (CD41) for single cells in each cluster. Cell clusters were labeled and colored by cell type names: 1, hematopoietic stem cells-1 (HSC-1); 2, HSC-2; 3, Pre-GMP; 4, GMP-1; 5, GMP-2; 6, GMP-3; 7, megakaryocyte-erythroid progenitor (MkE-P); 8, EryP-1; 9, EryP-2; 10, erythroid blast-1 (EryB-1); 11, EryB-2; 12, leukemia-1; 13, leukemia-2; 14, megakaryocyte progenitor-1 (MkP-1); 15, MkP-2; 16, megakaryocyte blast (MkB); and 17, B-cell progenitor (B-cell-P). Dot plot displays average gene expression of surface marker genes for identification of a cluster cell type. The color key from white to red indicates low to high average gene expression level, respectively. The dot size indicates percentage of cells expressing a certain marker. (C) Unsupervised graph-based clustering of scRNA-seq data set projected onto a t-distributed stochastic neighbor embedding (t-SNE) plot displaying distinct clusters. A total of 29 029 cells were clustered in the left panel. Each point represents an individual cell. Cell clusters were labeled and colored according to cell type (HSPC subpopulations and leukemic population). In the right panel, t-SNE plot displays subpopulations of HSPCs and leukemic population in a WT, JVF, a pre-PEL JVFP−/−, and a PEL JVFP−/− mouse. (D) Heat map of hierarchical clustering of Pearson correlation coefficients between clusters. All coefficients based on the average normalized scRNA-seq data in each cluster. (E) Enrichment analysis of Gene Ontology (GO) and pathways associated with upregulated genes in leukemia cells relative to erythroid progenitors and blasts. (F) Inferred CNA profiles from scRNA-seq data. Chromosomal gains (red) and deletion (blue) are inferred by averaging expression over 100-gene stretches on the respective chromosomes (columns) across the single cells (rows). Top: cells from a WT mouse as reference cells not expected to contain CNVs. Bottom: cells tested for CNAs relative to the reference cells. Zoom in of CNA profiles for leukemia, EryP, and EryB groups on chromosome 4, 6, 10, and 15 (right).

Close Modal
This Feature Is Available To Subscribers Only

or Create an Account

Close Modal
Close Modal